Regional Variations in Clinical Trial Outcomes in Oncology
Princess Margaret Hospital, University of Toronto, Toronto, Ontario, Canada;
Search for other papers by Brooke E. Wilson in
Current site
Google Scholar
PubMed
Menzies Centre for Health Policy, University of Sydney, Sydney, Australia.
Search for other papers by Sallie-Anne Pearson in
Current site
Google Scholar
PubMed
Search for other papers by Michael B. Barton in
Current site
Google Scholar
PubMed
Search for other papers by Eitan Amir in
Current site
Google Scholar
PubMed
Background: It is unknown how often regional differences in oncology trials are observed. Based on our study findings, we quantified regional variation in registration studies in oncology and developed a question guide to help clinicians evaluate regional differences. Methods: Using FDA archives, we identified registration studies in solid tumor malignancies from 2010 to 2020. We extracted the baseline study characteristics and participating countries and determined whether the primary publication reported a regional subgroup analysis. For studies presenting outcomes stratified by region, we extracted the stratified hazard ratios (HRs) and extracted or calculated the test for heterogeneity. We performed a random effects meta-analysis and a pairwise comparison to determine whether outcomes differed between high-income versus mixed-income regions. Results: We included 147 studies in our final analysis. Studies supporting FDA drug approval have become increasingly multinational over time (β = 0.5; P=.04). The median proportion of countries from high-income groups was 81.2% (range, 44%–100%), with no participation from low-income countries in our cohort. Regional subgroup analysis was presented for 78 studies (53%). Regional heterogeneity was found in 17.8% (8/45) and 18% (8/44) of studies presenting an overall survival (OS) and progression-free survival endpoint, respectively. After grouping regions by income level, we found no difference in OS outcomes in high-income regions compared with mixed-income regions (n=20; HR, 0.95; 95% CI, 0.84–1.07). To determine whether regional variation is genuine, clinicians should evaluate the data according to the following 5 questions: (1) Are the regional groupings logical? (2) Is the regional difference on an absolute or relative scale? (3) Is the regional difference consistent and plausible? (4) Is the regional difference statistically significant? (5) Is there a clinical explanation? Conclusions: As registration studies in oncology become increasingly international, regional variations in trial outcomes may be detected. The question guide herein will help clinicians determine whether regional variations are likely to be clinically meaningful or statistical anomalies.
Submitted January 3, 2022; final revision received April 21, 2022; accepted for publication May 9, 2022.
Author contributions: Study concept: Wilson. Data extraction: Wilson. Statistical analysis: Wilson, Amir. Data interpretation: All authors. Manuscript preparation: All authors.
Disclosures: Dr. Pearson has disclosed receiving grant/research support from Abbvie, Inc. Dr. Amir has disclosed receiving personal fees from Agendia BV, Apobiologix, Genentech, Inc./Roche Laboratories, Inc., Novartis Pharmaceuticals Corporation, and Sandoz. The remaining authors have disclosed that they have not received any financial consideration from any person or organization to support the preparation, analysis, results, or discussion of this article.
Funding: Dr. Wilson was supported as a National Breast Cancer Foundation of Australia International Fellow.
Supplementary Materials
- Supplemental Materials (PDF 632 KB)
© 2019-2024 National Comprehensive Cancer Network